Abstract
Internalism about a person’s good is roughly the view that in order for something to intrinsically enhance a person’s well-being, that person must be capable of caring about that thing. I argue in this paper that internalism about a person’s good should not be believed. Though many philosophers accept the view, Connie Rosati provides the most comprehensive case in favor of it. Her defense of the view consists mainly in offering five independent arguments to think that at least some form of internalism about one’s good is true. But I argue that, on closer inspection, not one of these arguments succeeds. The problems don’t end there, however. While Rosati offers good reasons to think that what she calls ‘two-tier internalism’ would be the best way to formulate the intuition behind internalism about one’s good, I argue that two-tier internalism is actually false. In particular, the problem is that no substantive theory of well-being is consistent with two-tier internalism. Accordingly, there is reason to think that even the best version of internalism about one’s good is in fact false. Thus, I conclude, the prospects for internalism about a person’s good do not look promising.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
In the view of L. W. Sumner, the correct theory of well-being must be internalist (or subjective, as he puts it) in the sense that it ‘make[s] your well-being depend on your own concerns: the things you care about, attach importance to, regard as mattering, and so on.’ (Sumner 1996, p. 42). Moreover, Connie Rosati mentions a long list of philosophers, including Rawls and Brandt, who hold substantive theories of welfare that are internalist in spirit (cf. Rosati 1996, p. 298, note 2). Furthermore, Darwall (1983, p. 81), Korsgaard (1986, p. 5) and Williams (1982, p. 106) endorse internalism about normative reasons, which, one suspects, would make them sympathetic to internalism specifically about a person’s good, as well.
Railton offers a bit of intuitive support for the view, alleging that it would be counter-intuitive to think that something that is ‘highly alien’ to a person could be part of that person’s good (cf. Railton2003, p. 47). Moreover, Velleman gives an overt argument for the view. In particular, he presents a version of the argument that Rosati calls the argument from ‘ought’ implies ‘can’ (cf. Velleman 2000, pp. 93–96; Rosati 1996, pp. 320–322).
Cf. Rosati, pp. 301–302.
Cf. Rosati, p. 303.
Rosati often uses ‘existence internalism’ to refer to internalism about a person’s good, in order to distinguish it from judgment internalism. I will continue to use the phrase ‘internalism about a person’s good’, however.
Cf. Rosati, p. 311.
Perhaps the argument would be more plausible if it relied on the premise that non-cognitivism and internalism about a person’s good are the only available philosophical explanations of the truth of judgment internalism. However, in that case first premise would have to be modified to state that ‘A philosophical explanation is needed of the truth of judgment internalism’, which may seem somewhat implausible. After all, one might wonder why only a philosophical explanation of the truth of judgment internalism would be adequate. Perhaps a psychological explanation might do as well.
Of course, some non-cognitivists have taken steps towards addressing this second concern by showing how some of the problematic features of anti-realism might be avoided (see, for instance, Blackburn 1993).
Could one save this premise by pointing out that judgment internalism does not have to be the strong claim that all sincere judgments about one’s good motivate, but could rather merely be the weaker claim that sincere judgments about one’s good normally motivate? This will not help. The same problem persists. After all, even under normal conditions, most of us will frequently make sincere but mistaken judgments about what is good for us. And in such cases (even though conditions are normal) two-tier internalism still will not guarantee that these mistaken judgments will motivate. Thus from the truth of two-tier internalism, not even this weaker version of judgment internalism (i.e. the claim that sincere judgments about one’s good normally motivate) would follow. (Thanks to Kristian Olsen for pressing me on this point.).
Note that this metaphysical argument is prefigured, but not explicitly developed, in Velleman (2000, p. 86).
Note that these arguments do not issue in the conclusion that two-tier internalism is true, merely that some form of internalism is true. Rosati says that ‘the argument also supports two-tier internalism, albeit more indirectly.’ (p. 315) But when she goes onto explain how this indirect argument is supposed to work, she admits that ‘this argument for [two-tier] internalism cannot be entirely independent of the first [i.e. the argument from judgment internalism]’ (p. 315). We have already seen that the Argument from Judgment Internalism fails, so I focus here only on the Argument from the Metaphysics of Value as it applies to internalism about a person’s good in general.
For a more precise account along these lines of ‘natural property’, see Copp (2004, pp. 12–13).
See Mackie (1977, chap. 1).
There are other ways to understand ‘motivational property,’ of course (e.g. as a property that is had by a creature’s motivational system). But it seems that if any of these alternative interpretations of ‘motivational property’ are adopted, internalism about a person’s good would not follow from the fact that goodness for a person is a motivational property. So premise 3) would become false.
Foot (2001) has defended an account of this sort.
Rosati mentions McDowell’s ‘sensibility theory’ as an example of a naturalistic account on which goodness for a person is not a motivational property in the sense of D1 (cf. p. 314). However, she still thinks that that the sensibility theory is an internalist theory. Thus the Metaphysical Argument could be reformulated so as to take account of the sensibility theory. However, the problem I raise in the main text would remain. In particular, this modified version of argument would still be undermined by the fact that there are some versions of naturalism about goodness for a person, like the Aristotelian account, on which internalism is not obviously true.
Another way to understand premise 1) in the Second Metaphysical Argument would be to take it to be the claim that the property of goodness for a person itself would not exist if there were no creatures for whom things can matter. I think this understanding of premise 1) is less plausible than the one mentioned above, since I am inclined think that properties can exist even if they are never instantiated. However, this alternative understanding of premise 1) would not save the argument. After all, there seems to be no reason to think that from the fact that the property of goodness for a person would not exist if there were no people around, it follows that goodness for a person in the actual world, where it does exist, is a motivational property. According to a metaphysics on which properties that are never instantiated do not exist, the property of being salami would not exist if sausage-making creatures did not exist. But why would it follow from this that the property of being salami in the actual world, where this property does exist, is a motivational property? There seems to be no reason to think this.
Perhaps Rosati could respond to this argument by pointing out that the property of being salami is an artifactual property. She may grant that it is a mistake to infer that a given property, P, is a motivational property from the fact that P would not be instantiated if there were no creatures around. After all, artifactual properties (like that of being salami) are like this too. But Rosati may nonetheless insist that one can legitimately infer that P is a motivational property from the fact that both i) P would not be instantiated if there were no creatures around and ii) P is a non-artifactual property. (Thanks to Scott Hill for pointing out this line of response to me.)
However, this too would be a mistaken inference. After all, consider the property of being hairy. Or a phenomenal property like that of appearing to someone to be bluish. Both of these are non-artifactual properties that would not be instantiated if there were no creatures around. Still, neither one is a motivational property. Thus this line of response will not save the argument.
It should be noted that Rosati presents a version of the epistemological argument specifically in favor of two-tier internalism (see Rosati, p. 319). The only significant difference between the epistemological argument for two-tier internalism and the epistemological argument for simple internalism is that premise 1) would have be modified to say something like this:
-
(1′) The claim that X is good for one cannot be justified unless there are some conditions, C, such that i) one would care about X under C and ii) this is a fact that one would care about under ordinary optimal conditions.
I will not discuss this version of the argument separately, however, because it should be clear that the same problems that undermine the original version of the argument (in favor of simple internalism) afflict the modified version of the argument (in favor of two-tier internalism) as well. For it too equivocates on the notion of justification.
-
However, I am still inclined to think that premise 1), even on this interpretation, is false. What people can be convinced of (like what they are motivated to do) seems to depend to a large extent on what they believe. If one explicitly believes that externalism is true (i.e. that certain things can be good for one even if one does not care about them), then it might well be possible to convince this person that X is good for her even if she cannot care about X. Thus premise 1), even on this interpretation of it, would be false.
Cf. Velleman (2000, pp. 85–98).
In the above passage, Rosati says ‘we can be prima facie obligated to care about something only if it is at least prima facie an option…’ (my italics). Thus one might think premise 2) really should say this instead:
2’) If P prima facie ought to care about X, then P prima face can care about X.
However, I do not think this is a charitable way to interpret the argument. For one thing, this notion of ‘prima facie can’ is obscure. What does it mean for one to be prima facie able to do something? Second, if one takes the second premise in the argument to be 2’), then to preserve the validity of the argument, the conclusion would have be that if X is good for P, then P prima facie can care about X. However, this conclusion does not amount to any version of internalism. For internalism is not formulated in terms of ‘prima facie can’. What’s more, a version of the epistemological argument that employs 2’) would still fall prey to the same problems as the version discussed in the body of the paper.
Also note that Rosati presents a version of this argument in favor of two-tier internalism as well. However, it suffers from the same problems that undermine the original version of the argument, in favor of simple internalism.
This point is argued in detail by Peter Graham in ‘Some Thoughts on Prima Facie Moral Obligation’ ms.
Cf. Ross (1930, see especially chap. 2, ‘What Makes Right Acts Right?’).
Thanks to Fred Feldman both for pointing out this way of formulating the argument and for offering helpful comments about its problems.
See, for example, Nettle (2005, p. 154). John Stuart Mill also recognized this phenomenon: ‘I never, indeed, wavered in the conviction that happiness is the test of all rules of conduct, and the end of life. But I now thought that this end was only to be attained by not making it the direct end. Those only are happy (I thought) who have their minds fixed on some object other than their own happiness; on the happiness of others, on the improvement of mankind, even on some art or pursuit, followed not as a means, but as itself an ideal end. Aiming thus at something else, they find happiness by the way.’ (Cf. Mill 1909, p. 94.)
Rosati attempts to extend this argument to cover two-tier internalism as well, but again I focus primarily on the basic version of the argument because the expanded version suffers from the same problems as the basic version.
In particular, it seems that something can suit the nature of a rational and autonomous person, even though this person cannot deliberately choose to pursue this thing. For instance, although (let’s suppose) I am a rational and autonomous person, it seems that I cannot deliberately choose to digest my food. Like it or not, this simply happens to the food I eat. Nonetheless, my digesting the food I eat is surely something that suits my nature to a high degree. Thus there may be counter-examples to premise 2).
Clearly it would suit the nature of a fish to a lesser degree to find itself out of water than to be located in the water. But beyond such obvious cases, how are we to decide in a systematic way whether something is well suited or poorly suited to a given creature’s nature? I suspect that we will have a strong intuition that X suits the nature of creature C if and only if we have a strong intuition that X is good for C. Thus I am not convinced that one can argue in a non-question-begging way from intuitions about what suits the nature of a creature to claims about what can or cannot be good for that creature.
I suspect that a problem similar to the one I raise for two-tier internalism will threaten most other formulations of ‘stronger internalism’ as well (i.e. a version that is restrictive enough to adequately capture the basic internalist intuition). I will not explicitly argue for this here, however. My discussion of the problem for two-tier internalism should be sufficient to indicate how the problem would carry over to the other strong versions of internalism on offer. Note, however, that the problem I raise for two-tier internalism will not afflict something as weak as simple internalism. After all, every theory of welfare is going to be consistent with simple internalism. Nonetheless, we saw before (in Sect. 1) that simple internalism was unsatisfactory for other reasons.
Note that two-tier internalism has some technical problems as well, in addition to the substantive problem I present in the body of the text. However, perhaps more work would yield a version of two-tier internalism that avoids these technical problems. (Thanks to Fred Feldman for pointing these problems out to me.)
What are these technical problems? The second condition of two-tier internalism is intended to pick out the ‘appropriate conditions,’ i.e. those in which your preferences would be relevant to determining your good. However, it does not seem that the view’s second ‘tier’ succeeds in picking out the ‘appropriate conditions.’ For one thing, suppose you have not given any thought to the question of which conditions are relevant for determining your good. In that case, there might be no counterfactual conditions, C, such that in ordinary optimal conditions you would care what your preferences would be in C. As a result, nothing would be good for you. Second, suppose you are deeply confused and believe that the ideal conditions for determining your good are conditions in which you have fasted for 10 days. You think that when you are delirious with hunger, you are afforded special insight into the nature of the universe so that your preferences are optimally attuned to your good. Suppose you would go on thinking this even if you were in ordinary optimal conditions. In this case, two-tier internalism would imply that the ‘appropriate conditions’ for you are conditions in which you are delirious with hunger. But intuitively this does not seem plausible. It is unlikely that conditions of extreme hunger are appropriate for determining what your good is—even if you would care about your preferences in these conditions, while in ordinary optimal conditions.
Rosati acknowledges this herself in saying that ‘[i]t is important to see that two-tier internalism most directly tests theories rather than alleged goods.’ (Rosati, p. 308).
Cf. Feldman (2004, pp. 49–50).
Just to be clear, Stoicus Jr. sometimes inadvertently has desires, but whenever he realizes it, he immediately loses the desires in question. (Accordingly, it’s not the case that Stoicus Jr. has a desire not to get his desires satisfied. This might lead to certain paradoxes that I would rather steer clear of. For a good discussion of these paradoxes, see Bradley 2007.).
ODS might also seem inferior to EDS because the former makes the welfare facts modally unstable in an unattractive way, while the latter does not. Let ‘SApple’ refer to the state consisting just of my eating a particular apple. In a possible world where I desire SApple to a certain degree, this state would have one welfare value for me, according to ODS, while that very same state of affairs would have a different welfare value for me in a different possible world where I desire SApple to a different degree. By contrast, EDS does not have this consequence. Since SApple is not an episode of desire satisfaction, EDS implies that this state by itself can have no welfare value for me (even in a world where I desire it). What would have welfare value for me, on this theory, is a larger state of affairs consisting of my desiring, to a certain degree d, to eat the apple and my doing so. And this larger state, according to EDS, will have the same welfare value for me—a value corresponding to d—no matter what circumstances it occurs in. Thus insofar as one thinks the moral facts are not contingent, one might have reason to prefer EDS over ODS.
This argument is similar in spirit to one that Bradley makes in a slightly different context (cf. Bradley 2009, p. 19).
References
Blackburn, S. (1993). Essays in quasi-realism. Oxford: Oxford University Press.
Bradley, B. (2007). A paradox for some theories of welfare. Philosophical Studies, 133, 45–53.
Bradley, B. (2009). Well-Being and Death. USA: Oxford University Press.
Brink, D. O. (1984). Moral realism and the sceptical arguments from disagreement and queerness. Australasian Journal of Philosophy, 62, 111–125.
Brink, D. O. (1989). Moral realism and the foundations of ethics. Cambridge: Cambridge University Press.
Copp, D. (2004). Three grades of normativity. In P. Shaber (Ed.), Naturalism and normativity. Frankfurt: Ontos-Verlag.
Darwall, S. (1983). Impartial reason. Ithaca: Cornell University Press.
Feldman, F. (2004). Pleasure and the good life: Concerning the nature, varieties, and plausibility of hedonism. Oxford: Clarendon Press.
Foot, P. (2001). Natural goodness. Oxford: Clarendon Press.
Kagan, S. (1998). Rethinking intrinsic value. The Journal of Ethics, 2(4), 277–297.
Korsgaard, C. (1983). Two distinctions in goodness. The Philosophical Review, 92(2), 169–195.
Korsgaard, C. M. (1986). Skepticism about practical reason. Journal of Philosophy, 83, 5–25.
Mackie, J. L. (1977). Ethics: inventing right and wrong. New York: Penguin Books.
Mill, J. S. (1909). Autobiography & essay on liberty. Edited by Charles W. Eliot. New York: P.F. Collier & Sons.
Nettle, D. (2005). Happiness: The science behind your smile. Oxford: Oxford University Press.
Railton, P. (2003). ‘Facts and values’, in facts, values, and norms: Essays toward a morality of consequence (pp. 43–84). Cambridge: Cambridge University Press.
Rosati, C. (1996). Internalism and the good for a person. Ethics: An International Journal of Social, Political and Legal Philosophy, 106(2), 297–326.
Ross, W. D. (1930). The right and the good. USA: Oxford Univeristy Press.
Sumner, L. W. (1996). Welfare, happiness, and ethics. New York: Clarendon Press.
Velleman, J. D. (2000). Is motivation internal to value? In The possibility of practical reason. USA: Oxford University Press.
Williams, B. (1982) Internal and external reasons. In Moral Luck. New York: Cambridge University Press.
Acknowledgements
I would like to thank Fred Feldman, Ernesto Garcia, Pete Graham, Scott Hill, Kristian Olsen, James Patten, Michael Rubin, Kelly Trogdon and an anonymous reviewer for this journal for their helpful comments on earlier drafts of this paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Sarch, A. Internalism about a person’s good: don’t believe it. Philos Stud 154, 161–184 (2011). https://doi.org/10.1007/s11098-010-9533-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-010-9533-0